Water Quality and Microbiological Contamination across the Fish Marketing Chain: A Case Study in the Peruvian Amazon (Lagoon Yarinacocha)

Abstract

The contamination of the surface water of lagoons is a common problem in developing countries, and can affect fishing activities. A case study was conducted on water quality and microbiological contamination of the fishing marketing chain in the Peruvian Amazon (Laguna de Yarinacocha). The microbiological, physical–chemical and parasitological parameters of the surface water were evaluated in three points of the lagoon near the landing stage; and microbiological parameters of facilities, handlers and three species of fish (Prochilodus nigricans, Mylossoma duriventre and Siluriforme spp.). In the water, there were coliform counts ≥ 23 (Most probable number—MPN)/100 mL, Escherichia coli ≥ 3.6 MPN/100 mL, and Pseudomona spp. up to 2.2 MPN/100 mL; high turbidity and variable amounts of parasites. In facilities and handlers, high levels of coliforms, mainly Escherichia coli, and Staphylococcus aureus and Escherichia coli, were found in M. duriventre meat. A poor quality of the surface water of the lagoon is concluded that compromises part of the fishing marketing chain, mainly facilities and manipulators. Furthermore, the levels of Staphylococcus aureus and Escherichia coli in fish meat show poor handling practices and possible risk of contamination by water sources.

1. Introduction

The surface water bodies of the lagoons and rivers in the Amazon are an essential source to maintain the existence of the population. People have preferentially settled near these sources for millennia, because their ecosystems are important from an economic and ecological perspective. However, the alteration of these resources has gone hand in hand with human activity [1,2]. Factors related to anthropogenic activities such as land use, the discharge of domestic waste from nearby cities or towns, the construction of dams and fishing pressures can negatively affect the quality of habitat and biodiversity, mainly due to the contamination of surface waters [3,4].

Water may be contaminated by several pathogens, including bacteria, viruses, protozoa, and helminthes. According to the World Health Organization (WHO, Geneva, Switzerland), 80% of all illnesses in developing countries are the result of contaminated water [5]. The risks of microbial contamination through bacterial pathogens are most common, including enteric and aquatic/environmental pathogens [6,7]. Enteric pathogens are mainly attributed to fecal contamination [8]. Water serves as a route of transmission for environmental pathogens to humans [9,10,11], allowing dissemination of bacteria towards some activities such as small-scale fishing or disembarkation, which are especially important in emerging countries [12].

Weather conditions in tropical countries, with high annual average rainfall, have reported high loads of Escherichia coli in surface waters of their small watersheds [13]. In Peru, the presence of biological contaminants, such as E. coli in two areas of surface water in the Sechura Bay in the northern coast of the country (2.4 CFU/100 mL and 1.6 CFU/100 mL), has negatively impacted the food safety of fish derived from that Bay [14]. Similarly, researchers have identified, Enterobacterales of the genus Citrobacter sp., Enterobacter sp., Klebsiella sp. Edwardsiella sp., Proteus sp., as well as bacteria of the genus Vibrio sp., Staphylococcus sp., and Streptococcus sp. in the water used for hydrobiological products in the markets of cities in other countries of the Andean region [15].

Water resources monitoring and the identification of sanitary risks are priorities in many emerging countries; as well as sanitary surveillance of fishery products in urban and rural areas. It is important to investigate the quality of natural water resources and how water quality might influence food safety [16]. Sanitary deficiencies in critical points, such as fish marketing, can result in over 70% of product being discarded [17], and ultimately this type of contamination can also put the human population health at risk.

The objectives of this investigation were to (i) evaluate the quality of the surface water of a lagoon in the Peruvian Amazon, (ii) evaluate the microbiological contamination of critical points along the fishing marketing chain, including facilities, and food handlers, and (iii) evaluate the sanitary quality of the fish traded. With this research, we hypothesize that the lagoon’s surface water quality has the potential to compromise the sanitary status of the downstream processing and marketing of fish and of the fish product ultimately sold for consumption.

2. Materials and Methods

2.1. Place and Time of the Study

The study was carried out in the Yarinacocha lagoon and its fishing landing site, located in the Yarinacocha district, Coronel Portillo province, Ucayali region in the Peruvian Amazon. This lagoon has its origin in a section of the Ucayali river isolated due to the diversion of its course, whose source of fresh water is surrounded by abundant vegetation. It is also characterized by having urban rural populations settled on its shores, whose economic activity is concentrated in tourist navigation, fishing and the transport of cargo and passengers. In addition, population growth has caused pollution problems in the lagoon due to the sewage that is discharged into it.

Before landing, the captured fish are washed at 150 m. from the shore of the lagoon. The boats are installed in areas close to the shore, which are usually cleaned by the fishermen with the water from the lagoon. The process ends with the sale of the fish in areas close to the shore. In addition, primary marketers occasionally use surface water from the lagoon to clean the surfaces of the vending areas (utensils, tubs, and tables) and their hands. When the process is finished, the solid waste is collected and transported in a truck to a final collection area located in the center of the city.

The surface water samples came from areas of the lagoon near the landing area (see Figure 1). They were also obtained from points near the commercialization area (facilities for the sale, handlers and fish for sale). The samples were processed in the Laboratory Diagnosis Unit of the Animal Health Section of the Veterinary Institute of Tropical and Altitude Research, Faculty of Veterinary Medicine, Universidad Mayor de San Marcos, Ucayali, Peru. The study was carried out during the rainy season between October 2017 and March 2018.

2.2. Sample Collection

The collection of samples was developed based on the critical points of mainly microbiological contamination, whose origin can be found in the water, which can act as a vehicle for cross contamination towards the entire fishing marketing chain developed at the landing stage of the lagoon (see Table 1).

Surface water samples were collected from the lagoon at three critical points related to the landing of fish: (1) located on the shore (0–1 m), (2) at 50 m from the shore and (3) at 150 m from shore (see Figure 1). One thousand milliliters of surface water was collected in sterile glass bottles for the microbiological and parasitological analysis, and in plastic bottles for the physical–chemical analysis. The sample collection was conducted according to methods described by Da silva et al. and The American Public Health Association [18,19]. The samples were stored between 2 and 8 °C and were processed within 2 h after collection.

For the landing and marketing facilities of the fishing activity, we sample 5 to 10 inert surfaces by swabbing the boats, retaining walls, sales tables and trays. Two samples were taken per surface, obtaining a total of 60 surfaces (30 for the determination of coliforms and Escherichia coli, and 30 for the determination of Salmonella spp.). Inert surface smears were performed according to the American Public Health Association (APHA) [20], supplemented by the Technical Guide for Microbiological Analysis and Surfaces in Contact with Food and Beverages [21]. To summarize briefly, a sterile swab was dipped into a tube with 10 mL of sterile peptone water and then rubbed over a 100 cm² (10 cm × 10 cm) area for sampling four times, limited by a sterile cardboard mold. The swab was pressed against the inside wall of the tube in a rotary motion to remove excess solution. Finally, the swab was placed in the tube, breaking and discarding the swab shaft. The samples were kept between 2 and 8 °C, until processing.

For the hands of the workers involved in the fishing market, samples were collected from 10 people involved in each point of the chain (handlers on landing, stevedores and primary traders). Thirty specimens from hand surfaces were collected (10 from handlers at landing, 10 stevedores and 10 primary marketers). We used the technique of microbiological evaluation of irregular surfaces (hand rinsing) described by the American Public Health Association (APHA) [20] and the Technical Guide for Microbiological Analysis and Surfaces in Contact with Food and Beverages [21]. Each person’s hands were immersed for 30 s in 100 mL of sterile peptone water collected in unused Ziploc bags. After removing the hands, the liquid was placed in a sterile 250 mL bottle. A sample was taken from both hands of each handler. The samples were kept between 2 to 8 °C until processing.

For the commercialized fish, a non-probabilistic convenience sampling was carried out, collecting 30 gutted fish from each of the three most commercialized species; “boquichico” Prochilodus nigricans, “palometa” Mylossoma duriventre and “catfish” Siluriforme spp. (five fish considered as a lot (chosen at random) of six stalls for each species, out of a total of 20 stalls at the landing stage) for microbiological analysis (total = 90). Sampling was carried out following the protocols of the National Fisheries Health Agency [22]. Additionally, six marketers (ten of each species) were sampled by the same method for parasitological analysis (total = 30).

Table 1. Types of samples and analysis of surface water and critical points in the commercialization of fisheries in the Yarinacocha landing site.

Critical Point (Collected Samples)	Analysis Type	Analysis Performed	Referential Method
Superficial water	Microbiological analysis	Total coliform count, E. coli and Pseudomona sp.	[18,23,24]
	Parasitological analysis	Count of free-living organisms such as protozoa, copepods, rotifers, nematodes in all their evolutionary stages	[19]
	Physical chemical analysis	Determination of pH, conductivity, hardness, iron and turbidity	[19]
Facilities and manipulators	Microbiological analysis of inert surfaces (vessels, retaining walls, vending tables and trays)	Coliform count, presence/absence of E. coli and Salmonella spp. in 100 cm2 of surface	[20,21]
	Microbiological analysis of living surfaces (hands of manipulators)	Coliform count Presence/absence of E. coli and Salmonella spp. in hands	[20,21,25]
Fish sold	Microbiological analysis of meat	Total mesophilic aerobic count E. coli count Staphylococcus aureus count Presence/absence of Salmonella spp.	[18,22]
	Parasitological analysis of meat	Discard parasitic forms visually and directly microscopically	[26]

Table 1. Types of samples and analysis of surface water and critical points in the commercialization of fisheries in the Yarinacocha landing site.

Critical Point (Collected Samples)	Analysis Type	Analysis Performed	Referential Method
Superficial water	Microbiological analysis	Total coliform count, E. coli and Pseudomona sp.	[18,23,24]
	Parasitological analysis	Count of free-living organisms such as protozoa, copepods, rotifers, nematodes in all their evolutionary stages	[19]
	Physical chemical analysis	Determination of pH, conductivity, hardness, iron and turbidity	[19]
Facilities and manipulators	Microbiological analysis of inert surfaces (vessels, retaining walls, vending tables and trays)	Coliform count, presence/absence of E. coli and Salmonella spp. in 100 cm2 of surface	[20,21]
	Microbiological analysis of living surfaces (hands of manipulators)	Coliform count Presence/absence of E. coli and Salmonella spp. in hands	[20,21,25]
Fish sold	Microbiological analysis of meat	Total mesophilic aerobic count E. coli count Staphylococcus aureus count Presence/absence of Salmonella spp.	[18,22]
	Parasitological analysis of meat	Discard parasitic forms visually and directly microscopically	[26]

2.3. Microbiological Analyses

The surface water samples were analyzed determining the most probable number (MPN) of total coliforms, Escherichia coli and Pseudomona sp., per 100 mL sample, specifying the confidence interval of possible values for each sample; as described by Da Silva et al., FDA and Villegas et al. [18,23,24] (see Table 1).

For the total coliform count, glass tubes with 10 mL of tryptose lauryl sulfate (LST) broth were inoculated with 10 mL of the surface water sample. Together these were incubated at 35 ± 0.5 °C for 24 to 48 h. Tubes with detectable growth (due to the presence of turbidity and gas) were sampled and inoculated in Brilliant Green Bile Broth (BGBB) and incubated at 35 ± 0.5 °C for 24 to 48 h to discard total coliforms; as well as inoculated in E. coli broth (EC) at 44 ± 0.2 °C for 24 h to discard thermotolerant coliforms. From the tubes positive for thermotolerant coliforms, a sample was plated on Eosin Methylene Blue agar (EMB) and incubated at 35 ± 0.5 °C for 24 ± 2 h to 48 ± 2 h. Colonies were then subjected to the indole test for confirmation of E. coli.

For Pseudomona sp., a 10 mL sample was inoculated in 10 tubes each with 10 mL of Asparagine broth and incubated at 35–37 ± 0.5 °C for 24 to 48 h. These isolates were further examined by the growth of colonies detectable by fluorescence under an ultraviolet (UV) lamp. A sample of the positive tubes was taken from the inoculated Asparagine broth and plated on Cetrimide agar at 35 ± 0.5 °C for 24 h. Colonies detectable by fluorescence were subjected to oxidase and catalase tests for confirmation of Pseudomona sp.

Samples from fishing landing and marketing facilities, as well as those from handlers in the marketing chain, were evaluated by counting the coliforms and determining the presence or absence Escherichia coli and Salmonella spp. according to procedures described by Da Silva et al., MINSA and Souza et al., with some minor modifications [20,21,25].

In the coliform count, the content of the first landing facility sample or the sample in the handwash bottle was homogenized manually. Then 0.2 mL was extracted from each and was plated on two plates with MacConkey Agar (MAC) by extension with a Drigalsky loop (0.1 mL per plate) and incubated at 35 ± 1 °C for 24 to 48 h. The number of coliforms in colony forming units (CFU)/cm² for facilities was calculated by multiplying the total CFU/mL of the suspension by the volume of diluent solution used in the sampling (10 mL) divided by the swabbed or sampled surface area (100 cm²). The calculation of CFU/hand was obtained by determining CFU/mL of suspension by the volume of diluent solution used in the sampling.

For the determination of E. coli, the landing facility sample and the hand rinse sample were homogenized manually using circular motions. Then, one ml of the installation sample was extracted to inoculate it in 9 mL of LST broth at 35° ± 0.5 °C, for 24 ± 2 h to 48 ± 2 h or 10 mL of the rinse bottle to sow 1 mL in each of the 10 tubes with 9 mL of LST broth at 35 ± 0.5 °C for 24 ± 2 h to 48 ± 2 h (presumptive test for detection of coliforms). A loop was collected from the positive tubes (with the presence of gas and turbidity), to be seeded in 10 mL. of EC broth at 44.5 ± 0.2 °C for 24 ± 2 h to 48 ± 2 h. (for the discard of thermotolerant coliforms). From the positive tubes, a loop was collected for plating on EMB agar at 35 ± 0.5 °C for 24 ± 2 h to 48 ± 2 h. for confirmation of the presence of E. coli. The obtained metallic green glossy black colonies were subjected to the indole test to confirm the presence of E. coli.

To determine the presence of Salmonella spp. the facility samples (in the tube) and the hand rinse samples (in the bottle) were incubated at 37 ± 1 °C for 18 ± 2 h. Then, a thread was taken to sow it in 10 mL of Rapapport broth at 41.5 ± 1 °C for 24 ± 3 h. From the tubes with the presence of turbidity, a layer was taken to be sown in Xylose Lysine Deoxylate (XLD) Agar at 37 ± 1 °C for 24 ± 3 h. Colonies suspected of being Salmonella spp. they were subjected to biochemical tests and antisera to confirm the presence.

The collected fish samples were evaluated by means of the bacteriological analysis of the meat, to discard mesophilic aerobes, E. coli, Staphylococcus aureus and Salmonella spp. (see Table 1).

2.4. Other Laboratory Tests

In order to identify and quantify evidence of parasites, we conducted parasitological analysis of the surface water, adapting the sedimentation method and direct microscopic observation described by SMEWW [21] (see Table 1). The collected water sample was kept at rest for 24 h at 2 to 8 °C for sedimentation. The supernatant was decanted until obtaining approximately 50 mL of the sample with sediment. This sediment and sample were homogenized by manual shaking and aliquoted, distributing 3 mL on 30 microscope slides. The slides were examined by microscopy at 10× and 40×. Finally, the total number of parasitic forms observed on all slides was multiplied to determine the total number of Free-living organisms as protozoans, copepods, rotifers, nematodes in all of their evolutionary states (N° org/L) (amount of the initial sample).

Additionally, the parasitological analysis of the fish was carried out using microscopic and direct visual observation techniques described by Conroy [26]. This consisted of the dissection of the portion of the middle trunk left side of each specimen, direct observation and the collection of six muscle portions (of approximately 0.2 cm²) with dissection scissors, for extension on microscope slides with the addition of a drop of physiological saline and observation at 10× to detect digenea or other parasitic forms. The observation was made with a Zeiss Standard 20 microscope (Germany).

Physical chemical analysis of the water was also carried out by measuring the parameters of pH, turbidity, hardness, conductivity and iron. The pH was determined by electrometric methods, using a commercial HANNA, HI2210, benchtop pH meter, stabilized and calibrated with standard solutions (pH 4, 7 or 10). Conductivity was determined using a HANNA HI9033 commercial multi-range portable conductivity meter. The water hardness and the iron concentration were determined by adapting the procedures described by SMEWW [21]. Finally, the turbidity was determined with a commercial HANNA portable turbidimeter, HI93703C, calibrated by means of standard solutions for its calibration and the measurement of the sample in Nephelometric Turbidity Units (NTU).

2.5. Statistical Analysis

Descriptive analysis was used and compared to Peruvian national standards of environmental water quality [27]. The quantity of coliforms in surface water samples were determined to be above or below the Maximum Permissible Limits (MPL) as defined by the country’s regulations [21]. The quantities of coliforms and other bacteria from fish meat samples were compared to Peruvian regulatory criteria for food safety of seafood [22].

3. Results

3.1. In the Surface Water

During the microbiological evaluation we found high levels of total coliforms (>23 MPN/100 mL), Escherichia coli [16 MPN/100 mL (5.9–33) from 0–1 m and 23 MPN/100 mL (8.1–53) at 50 m], and high levels of Pseudomona sp., especially at the furthest points [1.1 MPN/100 mL (0.05–5.9) at 50 m and 2.2 (0.37–8.1) at 150 m] (see

Table 2. Microbiological evaluation of the Surface water of the Yarinacocha lagoon, Ucayali.

Distance from the Disembarking Port (m)	Microbiological Analysis
	Total Coliforms (MPN/100 mL)	Escherichia coli (MPN/100 mL)	Pseudomona sp. (MPN/100 mL)
0–1	>23	16 (5.9–33)	<1.1
50	>23	23 (8.1–53)	1.1 (0.05–5.9)
150	23	3.6 (0.91–9.7)	2.2 (0.37–8.1)

MPN: Most Probable Number of Colony Forming Units per 100 mL of water sample. Bold numbers indicate values outside the recommended parameters. The values in parentheses represent the range of possible values in the MPN/100 mL for the result obtained from the sample, according to the methodology used. E. coli MPL (MNP/100 mL) for surface water for the production of drinking water and recreation (includes washing food for sale) (absent), surface water for extraction and cultivation of hydrobiological species in lakes or ponds (not established). Values of the other parameters within normal limits. Compared with reference values [27].

). The physical–chemical parameters showed high turbidity at the points nearest to the shore (53.39 NTU from 0–1 m and 305.6 NTU at 50 m) (see

Table 3. Physical–chemical evaluation of the Surface water of the Yarinacocha lagoon, Ucayali.

Distance from the Disembarking Port (m)	Parameters
	pH	Conductivity (µs/cm)	Hardness (mg CaCO3/L)	Iron (mg Fe++/L)	Turbidity (UNT)
0–1	7.49	282	120	0.34	53.39
50	8.1	183.1	88	0.3	305.6
150	7.9	204.5	104	<0.06	32.6

NTU: Nephelometric Turbidity Unit. Bold numbers indicate values outside the recommended parameters. Turbidity MPL (UNT or mg/L) for surface water for the production of drinking water (includes washing food for sale) (≤1000–1500), recreation water (≤100) and surface water for extraction and cultivation of hydrobiological species in lakes or ponds (not established), extraction of hydrobiological products under other conditions (≤60–80). Values of the other parameters within normal limits. Compared with reference values [27].

). During the parasitological evaluation, we found a large amount of free-living organisms (4600 org/L) at the nearest point of the evaluation (0 to 1 m of distance) (see

Table 4. Parasitological evaluation of the surface water of the Yarinacocha lagoon, Ucayali.

Source	Distance (m) from the Shore	Parasitological analysis (N° org/L)
Yarinacocha disembarking	0–1	4600
	50	1225
	150	0

).

3.2. In the Facilities and Handlers for Fish Marketing

In the facilities, surfaces and hands of the manipulators, we found high counts of coliforms and frequent presence of Escherichia coli. In addition, the highest bacterial loads in surfaces were found on the vending tables (80%, ≥1 CFU of coliforms/cm²) and on the stevedores’ hands (100%, ≥100 CFU of coliforms/hands and E. coli). It is worth noting the absence of Salmonella spp. in all of the evaluated samples (see

Table 5. Percentage of samples with coliforms ≥ Maximum Allowable Limit (LMP) and presence of Escherichia coli from the points of the fishing marketing chain at the Yarinacocha landing site.

Critical Point in the Fish Marketing Chain		Samples with	Samples with
		Coliforms ≥ LMP * (%)	Presence of E. coli (%)
	Boats	40	40
Facilities and furniture	Retaining walls	40	20
	Vending tables	80	20
	Trays	70	40
	Handler at disembarkation	80	80
Handlers’ hands	Stevedores	100	100
	Primary marketers	60	60

It was evaluated; the coliform count in facilities (CFU/cm² of surface), presence/absence of E. coli in 100 cm² of surface in boats, facilities and workplace surfaces, count of coliforms in the hands of handlers (CFU/hand) and presence/absence of E. coli in the hands of handlers. * Coliforms can include genera such as Citrobacter sp., Enterobacter sp., Klebsiella sp., Edwardsiella sp. and Proteus sp. Compared with reference values [21].

).

3.3. In Commercialized Fish

Mesophilic bacteria and Escherichia coli were present, but below the Maximum Permissible Limit (MPL); only Staphylococcus aureus was present above the MPL in the meat of the palometa; while there was no presence of Salmonella spp. in all samples (

Table 6. Microbiological evaluation of the meat of three hydrobiological species sold at the Yarinacocha lagoon landing stage.

Microorganism	Boquichico (Prochilodus nigricans)		Palometa (Mylossoma duriventre)		Bagre (Siluriforme sp.)		Maximum Permissible Limit/g
	n	c	n	c	n	c
Total Aerobic Mesophyll (CFU/g)	5	0	5	0	5	0	m (5 × 105), M (106) (n:5, c:2)
Escherichia coli (CFU/g)	5	2	5	1	5	0	m (10), M (102) (n:5, c:3)
Staphylococcus aureus (CFU/g)	5	1	5	3	5	2	m (102), M (103) (n:5, c:2)
Salmonella spp.	5	0	5	0	5	0	Absence in (n:5, c:0)

The bold values indicate that they have exceeded the maximum permitted limits (LMP) by the Sanitary Technical Standard. CFU: Colony Forming Units; n: number of sample units selected randomly from a batch. c: maximum number of sample units that can contain a number of microorganisms between “m” and “M” in a sampling plan. If a number of sample units is higher than “c”, the batch is rejected. m: microbiological limit that divides the acceptable quality from the rejectable one. M: values with microbial counts higher to M are unacceptable; the food constitutes a health Hazard. In pathogens such as Salmonella sp., the presence or absence in 25 g of meat sample was considered. Values in bold, demonstrate unacceptable values from the sample lot. Compared with reference values [22].

). Finally, no parasitic forms were found in the meat samples.

4. Discussion

In summary, from the results obtained we can evidence the existence of substandard surface water quality in this Yarinacocha lagoon, in addition to high levels of microbiological contaminants in the landing facilities and the handlers and moderate levels in the fish sold; according to the comparison with the sanitary technical standards [21,22,27] (see Figure 2).

The results obtained in the analysis of the surface water sample agree with those obtained by the Environmental Assessment and Enforcement Agency—Perú [28] (Table 2). They found high levels of contamination with E. coli and thermotolerant coliforms were found in multiple points of the surface water of the Ucayali River. This supports existing literature that E. coli is an important indicator of fecal contamination in surface waters, leading to the possible presence of other enteric pathogens that may have an influence in the bad health conditions during food handling [29].

Our findings highlight the suboptimal health conditions in the fishing marketing chain at the landing port of the Ucayali river lagoon. One possible explanation for these microbiological findings is the lack of a wastewater treatment system in the city [30]. With wastewater outlets in areas near the lagoon, the carry-over of pollutants by the rains of other anthropogenic activities, and the presence of animals in the surrounding areas can be the cause of the contamination of surface waters, as mentioned by other authors [31]; however, commercialized fish were not greatly affected by this water contamination, in accordance with what was found by Ramos—Ramirez [32].

Pathogenic bacteria are generally spread through water contaminated with human or cattle feces [33] and can result from human, industrial and farming activities of the area. The presence of other microbial agents in water, which compromise public health, can be found together with E. coli [34,35], such as high amounts of Pseudomona sp. [36]. These and other microbes have been implicated in causing diarrheal diseases due to water contamination in the north of the Peruvian Amazon. It is possible that the source of E. coli attributing to it the increase in disease outbreaks in the population [37]; however, these findings could be additionally associated with contamination during food handling, and not necessarily from the contamination of the meat of fish [7].

Most of the physical–chemical parameters were within normal ranges, only the turbidity levels were high (Table 3), compared to the environmental quality standards of water for port activities in the country [27]. In turbid systems, the role of suspended particles along with other environmental factors may be of great importance to promote the growth and distribution of pathogenic bacteria [38]. Besides that, comparatively large quantities of organic matter and turbidity favor the survival of fecal bacteria [39]. It should be noted that the turbidity may also be an important trigger for a greater microbial contamination for suspended solids [40], as found in the water samples of the areas nearest to the shore.

Turbidity conditions may be related to the high levels of free organisms in the water samples of the evaluated points nearest to the disembarking port (Table 4). Turbidity is strongly suggestive of eutrophication, which improves the nutrient availability and consequently the distribution of environmental pathogens [41,42]. In addition, the Ucayali River effluents discharged to the lagoon may contain fecal pathogens such as viruses, bacteria and protozoa. These microorganisms may be transported at significant distances [32,43]. Consequently, the contamination of lake water with coliforms including E. coli can represent a great sanitary risk for the collection of hydrobiological products [44].

The increase in turbidity suggests mainly the entrainment of solid waste in the surface water, which may cause a greater probability of surface contamination [45]. Furthermore, coliforms and E. coli have also been reported in other places such as lakes in the North Tropic of America, during rain events, as in this investigation [2]. Microbial contamination has also been reported in lake water in Malaysia in the Eastern Tropics [4]. Frequent rains in tropical countries could affect rural areas, especially areas with bare soil and/or annual crops, (since they lead to soil erosion and dispersal of any fecal matter on its surface), and also to urban areas, due to the discharge of sewage [46].

In the fishing marketing chain, the increased microbial contamination in the facilities, furniture and handlers (Table 5), suggest deficiencies in sanitary food handling practices. Inadequate hygienic practices in port sales can lead to cross contamination of products for food consumption [14]. In addition, the presence of the hands of fishermen exposed to coliforms and E. coli, which interact with resident fish, through skin mucus, may represent a risk at other levels [47]. These risks may interact with other human activities in the environment, although the role it may play in humans is not known [48].

The levels of Staphylococcus aureus above the MPL we found in Palometa meat (Table 5) are similar to those found by Vásquez [49], who determined the presence of Staphylococcus aureus over the permitted level in shellfish from several markets of a nearby city. Some pathogens can be eliminated thorough cooking; however, the common local tradition of eating raw, marinated or lightly cooked fish does not guarantee its elimination from the fish. In addition to these bacteria, they may be naturally present in the raw materials or directly added by human activity when handling practices are inadequate [50].

Although, the levels of E. coli were low, its presence in the meat of the fish evaluated, merits the need for periodic monitoring of the fish products. On the other hand, the absence of Salmonella spp. (Table 6) proves that this bacterium is not a biological fish contaminant, but it may be introduced in the marketing chain due to the compromised water quality or inadequate handling of the hydro biological products [51]. There are several studies about other types of meat, but there are few available reports on Salmonella spp. in hydro biological products [52].

The extent of the contamination found in our study could be similar to a study carried out in a lagoon in Zimbabwe, Africa, where the risk of fish consumption was described [53]. A similar study was conducted examining chemical and bacteriological contamination in a lake in Junín, Peru, however, they identified a widely compromised aquatic ecosystem [54]. This work can also be compared to the moderate contamination of the water of a lagoon in Tingo María in Huánuco, Peru, where several activities were restricted to avoid the increase of the contamination [55].

The findings of this research suggest an exposure of the critical points of the fish marketing chain to the substandard quality of the surface water in the lagoon; however, contamination in commercialized fish was low (Figure 2). The development of studies on the influence of poor surface water quality in all activities that guarantee food safety, such as fish marketing, is very important. Likewise, these will allow traceability to know the origin of the contamination at the critical points of the commercialization process [47].

The collection of samples from the facilities as well as from the handlers in the early morning hours in which the fishing landing in the lagoon takes place, the difficulty for the conservation and transport of the samples, and their processing with microbiological culture techniques; were some limitations for the development of this investigation. On the other hand, although the presence of E. coli in the evaluated points may include pathogenic and non-pathogenic strains, the typification of the strains could have given a greater impact to this research, to know the risks for public health. A genotyping of E. coli isolates to determine the presence of pathogenic strains, such as E. coli O157:H7, and the relationship between the strains isolated in surface water and the points of the fish marketing chain in the lagoon, could elucidate the source of contamination in final products.

In particular, the implementation of a constant application of good handling and management practices in the provision of drinking water for the hygiene of the facilities, furniture and manipulators; it could reduce the contamination of surfaces and fish as a final product. In general, investment in infrastructure is required to improve sanitary conditions in the fishing marketing chain and greater control of anthropogenic activities that compromise the quality of surface waters in the Yarinacocha lagoon. Finally, the implementation of disinfection and sanitation programs, as described in [56], and a Threat Analysis and Critical Control Points (HACCP), of the surface water sources where this activity takes place, as described in [57]. They will allow a better implementation of the operational sanitation programs for the commercialization of fishery products.

5. Conclusions

The poor quality of the surface water of the Yarinacocha lagoon compromises part of the fishing marketing chain, mainly facilities and manipulators. Furthermore, the levels of Staphylococcus aureus and Escherichia coli in fish meat evidence poor handling practices and possible risk of contamination by water sources.

The implementation of good handling practices and the supply of permanent drinking water throughout the marketing chain; In addition, the improvement of the landing stage infrastructure could reduce the contamination of surfaces and fish.